I just added the black outline and white-filled versions of the Flink logo to
the svn directory[1] for the Apache wide logo directory. I got these from the
presumably official directory[2] on the web site. This means that there are now
flink-[123].svg forms in that directory which should show
Ted Dunning created FLINK-4793:
--
Summary: Using a local method with :: notation in Java 8 causes
index out of bounds
Key: FLINK-4793
URL: https://issues.apache.org/jira/browse/FLINK-4793
Project: Flink
A big part of the cost of currentTimeMillis() is that it does a lot of work
to make sure that the time is really time.
If you only need a monotonic timer, nanoTime() might be what you want
instead of currentTimeMillis(). nanoTime() is particularly handy when you
want to avoid issues to do with le
Out of order is ubiquitous in the real-world. Typically, what happens is
that businesses will declare a maximum allowable delay for delayed
transactions and will commit to results when that delay is reached.
Transactions that arrive later than this cutoff are collected specially as
corrections whi
On Sat, Jun 20, 2015 at 12:52 AM, Stephan Ewen wrote:
> There is no comment section, otherwise
> I would have added that...
>
Email to the author. Better coverage next time.
http://www.datanami.com/2015/06/12/8-new-big-data-projects-to-watch/
l is
> proposing so I'm reluctant to comment specifically. But the bottom line is
> still the bottom line: LICENSE and NOTICE must reflect the bundled bits.
>
> Hope this helps,
>
> Marvin
>
> On Fri, Jun 12, 2015 at 9:45 AM, Ted Dunning
> wrote:
> > Marvin,
&
On Mon, Jun 15, 2015 at 8:45 AM, Maximilian Michels wrote:
> Just to give an idea what null values could cause in Flink: DataSet.count()
> returns the number of elements of all values in a Dataset (null or not)
> while #834 would ignore null values and aggregate the DataSet without them.
>
Compa
The example of SQL is obviously dominating thoughts of NULL, but I think
that the example of R is probably better in terms of how things can work
fairly well. NULL is a key concept and very helpful in a number of
settings. With R's fairly simple functional nature it is easy to filter
data and mos
gy. I’ll open a JIRA for that.
> >
> > Maybe also the default learning rate of 0.1 is set too high.
> >
> >
> > On Thu, Jun 4, 2015 at 1:20 AM, Ted Dunning
> wrote:
> >
> >> Any form of generalized linear regression should use adaptive learning
> &g
On Thu, Jun 4, 2015 at 1:26 PM, Till Rohrmann wrote:
> Maybe also the default learning rate of 0.1 is set too high.
>
Could be.
But grid search on learning rate is pretty standard practice. Running
multiple learning engines at the same time with different learning rates is
pretty plausible.
Al
Any form of generalized linear regression should use adaptive learning
rates rather than simple SGD. One of the current best methods is adagrad
although there are variants such as RMS prop and adadelta. All are pretty
easy to implement.
Here is some visualization of various methods that provides
It would also be relatively simple (I think) to retarget drill to Flink if
Flink doesn't provide enough typing meta-data to do traditional SQL.
On Tue, May 26, 2015 at 12:52 PM, Fabian Hueske wrote:
> Hi,
>
> Flink's Table API is pretty close to what SQL provides. IMO, the best
> approach woul
If there is an active twitter entity for Flink, I recommend framing that on
the home page.
On Mon, May 11, 2015 at 8:51 AM, Ufuk Celebi wrote:
> Hey all,
>
> I reworked the project website the last couple of days and would like to
> share the preview:
>
> http://uce.github.io/flink-web/
>
> I
Henry,
I think that the requirement is that project decisions need to be made on
an Apache controlled resource.
Not sure how many project decisions are made on the user list. I know I
answer a ton of Mahout questions on quora.
Having more than one arena for answers and reflecting between them f
Somebody moderated this email through to this list. That was an error.
Let's be more accurate here.
On Mon, Mar 16, 2015 at 6:31 AM, E-ZPass Agent wrote:
> Dear Dev,
>
> You have not paid for driving on a toll road.
> You are kindly asked to pay your debt as soon as possible.
>
> The copy of
On Mon, Mar 2, 2015 at 5:17 PM, Stephan Ewen wrote:
> There are two execution modes for the runtime: "reuse" and "non-reuse".
That makes a fair bit of sense.
This is going to have profound performance implications if this is the only
path for iteration.
On Fri, Feb 27, 2015 at 10:58 PM, Stephan Ewen wrote:
> I vote to have the key extractor return a new value each time. That means
> that objects are not reused everywhere where it is possible, but s
As the community of flink add-ons grows, a CPAN or maven-like mechanism
might be a nice option. That would let people download and install
extensions very fluidly.
The argument for making Apache contributions is definitely valid, but the
argument for the agility of fostering independent projects
On Thu, Jan 15, 2015 at 10:18 AM, Henry Saputra
wrote:
> I do use Git rebase before pushing my own PRs.
>
> So, no one uses the script to merge PRs?
>
> I will send update to the script to fix it.
>
Manual rebase is a good standard of practice. I like looking in detail at
what I am doing.
On Thu, Jan 8, 2015 at 5:09 PM, Henry Saputra
wrote:
> I am trying to hook us up with H2O guys, lets hope it pays off =)
>
I know the CEO and CTO reasonably well and will ping them.
Has anybody looked at NOTICE and similar files?
Sent from my iPhone
> On Jan 13, 2015, at 14:17, Stephan Ewen wrote:
>
> I am still traveling and could not yet run cluster tests, so I will delay
> my final vote till tomorrow.
>
> So far I can say:
>
> - All files have valid headers
>
> - Li
Argh. Some days I type faster than I read.
Sent from my iPhone
> On Jan 13, 2015, at 14:17, Stephan Ewen wrote:
>
> License and notice files are good for source and binary release
Would the user list do for now?
On Fri, Jan 9, 2015 at 7:27 AM, Robert Metzger wrote:
> Our PMC Chair or a ASF member has to request a list:
> https://infra.apache.org/officers/mlreq
>
> On Fri, Jan 9, 2015 at 4:25 PM, Kostas Tzoumas
> wrote:
>
> > Cool, does anyone know how we can do that wi
24 matches
Mail list logo
